Cisco Router 7200 bandwidth problem

[matt]

Hello
I would like to monitor bandwidth usage on my Cisco 7200 router. I added new host and services using host creator from Nagios XI console (i chose "Network switch) . Unfortunately when i open bandwidth graph there`s a "No performance graphs were found for this service." instead of the graphic. When i look in the nagios log file there`s an information:

[1292498296] SERVICE ALERT: 7200_ROUTER;Port 2 Bandwidth;WARNING;SOFT;4;(null)
or
[1292498656] SERVICE NOTIFICATION: nagiosadmin;7200_ROUTER;Port 2 Bandwidth;OK;xi_service_notification_handler;OK - Current BW in: Mbps Out: Mbps
As You can see above there`s no BW data received from the router.
I use VMWare trial appliance.
Thanks for any suggestions.
Matt

[tonyyarusso]

Could you please look under the "Advanced" tab and post what is displayed on the line for "Performance Data"? If it's blank, could you try running the check command from the shell of the server and paste the output? (I can tell you how to go about that if you aren't sure.)

[matt]

Hello
Here`s the output of "Advanced" tab

Service State: Ok
Duration: 1h 8m 55s
State Type: Hard
Current Check: 1 of 5
Last Check: 2010-12-17 06:55:13
Next Check: 2010-12-17 07:00:13
Last State Change: 2010-12-17 05:50:13
Last Notification: Never
Check Type: Active
Check Latency: 0.236 seconds
Execution Time: 0.0519 seconds
State Change: 8.02632%
Performance Data: in=Mb/s;4;10 out=Mb/s;4;10

regarding Your sugestion, how can i run "check" command ? Is this a some script etc?
Thx

[mguthrie]

Here's some useful info to check whether or not your performance data can produce a graph or not. If there's a syntax issue, or the performance data isn't in a valid format, PNP won't be able to generate a graph.
http://docs.pnp4nagios.org/pnp-0.4/abou ... quirements

To test a check command from the commandline, use a shell to access the check plugins directory:

cd /usr/local/nagios/libexec

Here's a simple example of a raw check being run from the command-line:
./check_icmp -H www.google.com

You'll need to identify what command is being run for your service, then look up the actual command-line arguments in the Configure->Core Config Manager->Commands page. Let us know if you need more details on this.

[matt]

Hello
Thank You for Your reply. I checked ./check_icmp command against www.google.com and received this string:

OK - www.google.com: rta 49,784ms, lost 0%|rta=49,784ms;200,000;500,000;0; pl=0%;40;80;;

I tried also run check_rrdtraf command against my router and here`s the result:

./check_rrdtraf -f /var/lib/mrtg/ROUTER_IP.rrd -w 20,20 -c 50,50 -l M
./check_rrdtraf: line 304: 171624,058600: value too great for base (error token is "058600")

[tonyyarusso]

Hmm, that's a new one for me. I'll see if I can find someone more familiar with the check_rrdtraf plugin to help figure this out.

[matt]

I`m waiting impatiently for any feedback. This is critical issue for us. We have to find a solution before buying it.

[mguthrie]

The error code that you're seeing looks like a bug or error in the check plugin itself. The fact that it's giving a line number tells where the script encountered an error. Based on what the error code is saying, the return value it's getting may have too many digits. Most of the strongly-typed programming languages make you specify the size of number that it's expecting. I'm not familiar with that particular plugin, but is there a way to change the scale of the performance data, for example, set an option to specify the performance data in seconds instead of milliseconds?

[admin]

Looks like it could be a bash parsing error. Bash is thinking the second number is octal when its not. I originally discovered this possibility when finding this post:

http://ubuntuforums.org/showthread.php?p=4206419

Tony - can you try and hack a fixed version of the check_rrdtraf plugin tomorrow?

[tonyyarusso]

Okay, I tried changing the particular line it was complaining about - give this file a shot and let us know if it works for you.